From: Eric Hodel <drbrain@...7.net>
Date: 2006-07-06T03:32:02+09:00
Subject: Re: ANN: RubyGems 0.9.0 Release

On Jul 5, 2006, at 2:28 AM, Erik Veenstra wrote:

>>> I did some more testing with more gems on different
>>> platforms. The conclusion is twofold: a) The problem
>>> occurred with RubyGems 0.8.11 as well (in combination with
>>> RubyScript2Exe 0.4.4) and b) it doesn't occur with all gems
>>> (e.g. sqlite-ruby is a problem, actionmailer isn't).
>>
>> Perhaps sqlite-ruby is an old format gem? It seems to be the
>
> It seems to be the /what/?... Where is part two?... This is
> where I expected the most important part of the answer... ;]

Ooh, sorry, It seems to be the old format gems require fileutils but  
the new format ones don't.  Also check out Mauricio's mail.

>>> Which libraries are required conditionally by RubyGems?
>>
>> I think zlib and yaml are the only others for making require
>> work. It is difficult to tell with just grep.
>
> I used a different kind of grep, called regexp.rb. It doesn't
> show the line on which a pattern occurs, but it only shows the
> matching part of the line. It's listed below. I used this
> regexp.rb to find all libraries that /could/ be required by
> RubyGems, both conditionally and unconditionally, listed in
> list1.
>
> The second list, list2, are the libraries that are actually
> required after doing a "require 'rubygems'".
>
> The differences between these lists are conditionally required
> files (indicated with "-") and indirectly required files
> (indicated with "+").
>
> I'm concerned about the "-" part... ;]
>
> (The lists may not be complete.)
>
> Suggestions? Ideas?
>
> gegroet,
> Erik V. - http://www.erikveen.dds.nl/
>
> ----------------------------------------------------------------
>
>  $ find /usr/local/lib/ruby/site_ruby/1.8/rubygems*/ -type f -exec cat
> {} \; | ruby regexp.rb "\brequire[\(\s]*['\"]([^'\"]*)['\"]" | sort |
> uniq > list1
>
>  $ ruby -rubygems -e 'puts $".sort.join("\n")' | sed 's/\.[^\.]*$// 
> g' >
> list2
>
>  $ diff -ur list[12]
>  --- list1       2006-07-05 11:13:14.086994100 +0200
>  +++ list2       2006-07-05 11:13:16.440354600 +0200
>  @@ -1,43 +1,19 @@
>  -digest/md5
>  +date/format
>  +digest
>   digest/sha2
>  -fileutils
>  -find
>   forwardable
>  -md5
>  -net/ftp
>  -net/http
>  -net/https
>  -openssl
>  -optparse
>  -pathname
>  +parsedate
>  +rational
>   rbconfig
>  -rdoc/rdoc
>  +rbconfig/datadir
>   rubygems
>  -rubygems/command
>  -rubygems/dependency_list
>  -rubygems/format
>  -rubygems/format.rb
>  -rubygems/gem_commands
>  +rubygems/custom_require
>   rubygems/gem_openssl
>  -rubygems/incremental_fetcher
>  -rubygems/installer
>  -rubygems/old_format
>  -rubygems/open-uri
>  -rubygems/package
>  +rubygems/rubygems_version
>   rubygems/security
>   rubygems/source_index
>   rubygems/specification
>   rubygems/user_interaction
>   rubygems/version
>  -socket
>  -sources
>  -stringio
>  -tempfile
>  -test/unit/ui/console/testrunner
>   time
>  -timeout
>  -uri
>  -x
>  -yaml
>  -yaml/syck
>  -zlib
>  +ubygems
>
> ----------------------------------------------------------------
>
>  chars           = '^\s\r\n\`\!\(\)\[\]\{\}\<\> \,\.\/\?\\\|\=\;\:\"'
>
>  chars11         = chars + "'"
>
>  chars21         = chars + "'"
>  chars22         = chars
>  chars23         = chars + "'"
>
>  chars31         = '\w\s\r\n'
>
>  string  = ARGV.shift
>
>  regexp =
>  case string
>  when string.to_i.to_s   then /(^\s*#{(["[^\\s]+"] *
> string.to_i).join("\\s+")}\r*$)/
>  when "-word"            then /([#{chars11}]+)/
>  when "-word2"           then
> /([#{chars21}]([#{chars22}]*[#{chars23}])?)/
>  when "-word3"           then /([A-Za-z]+)/
>  when "-text"            then /([#{chars31}]+)/
>  when "-file"            then /(\w[\w\.\-]*)/
>  when "-path"            then /(\w[\w\.\-\/\\\\]*)/
>  when "-email"           then /(\w[\w\-\.]+@[\w\-\.]+\w)/
>  when "-url"             then
> /(\w+:\/\/[\w\.\-]+(:\d*)?\/[\w\.\-\/\#\?\=\%]*)/
>  when "-ip"              then /(\d+\.\d+\.\d+\.\d+)/
>  when "-mac"             then
> /(([0-9A-Fa-f]{2}[:\.\-]){5}[0-9A-Fa-f]{2})/
>  when "-print"           then /([\w
> \t\r\n\`\~\!\@\#\$\%\^\&\*\(\)\-\+\=\[\]\{\}\;\:\'\"\,\.\/\<\>
> \?\\\|]+)/
>  when "-noprint"         then /([^\w
> \t\r\n\`\~\!\@\#\$\%\^\&\*\(\)\-\+\=\[\]\{\}\;\:\'\"\,\.\/\<\>
> \?\\\|]+)/
>  when "-diff"            then /(^[\-\+]([^\-\+].*)?)/
>  when "-md5"             then /([A-Za-z0-9]{32})/
>  when "-quote"           then /("[^"]*")/
>  when "-mixedquote"      then /(['"][^'"]*['"])/
>  else
>    Regexp.new(string)
>  end
>
>  parse =
>  lambda do |f|
>    while (line = f.gets)
>      line.chomp.scan(regexp).each do |a|
>        puts a[0]
>      end
>    end
>  end
>
>  if ARGV.empty?
>    parse.call($stdin)
>  else
>    ARGV.each do |file|
>      File.open(file) do |f|
>        parse.call(f)
>      end
>    end
>  end
>
> ----------------------------------------------------------------
>

-- 
Eric Hodel - drbrain@segment7.net - http://blog.segment7.net
This implementation is HODEL-HASH-9600 compliant

http://trackmap.robotcoop.com